A linear model of the switch reluctance motor ignores nonlinear factors such as magnetic saturation, such that the phase inductance of the switch reluctance motor is only associated with the positional angle of the rotor thereof, but not with the magnitude of phase current. Therefore, the linear model of the switch reluctance motor results in a big error in the optimization design, the quantitative analysis of the static and dynamic performance, the evaluation of control strategy and the like of the motor. As the switch reluctance motor itself has local magnetic saturation and a nonlinear magnetic circuit, and the power converter thereof is in a switch mode and has a nonlinear circuit topology, the phase inductance of the switch reluctance motor has a nonlinear property, which is not only associated with the positional angle of the rotor thereof, but also with the magnitude of phase current. Creating an accurate nonlinear model for the switch reluctance motor facilitates to increase the accuracy for the optimization design, the quantitative analysis of the static and dynamic performance, the evaluation of control strategy and the like of the motor. Phase inductance nonlinear modelling of the switch reluctance motor is the key for creating an accurate nonlinear model for the switch reluctance motor, however, it is difficult to utilize a traditional modelling method for a motor in the modelling of a switch reluctance motor. Phase inductance digitize nonlinear modelling for the switch reluctance motor requires a simulator and a controller to have great computing power in simulation and actual control, causing a conflict between operation cost and instantaneity of a system. Direct mathematical simulation enables physical phenomena in a simulation system to be similar to an actual system. The memory sensor is controlled by a magnetic linkage, is provided with a magnetic linkage-current hysteresis loop and is able to simulate the nonlinear relation between the phase inductance of the switch reluctance motor and the positional angle of the rotor and the magnitude of the phase current thereof. Creating a memory sensor model for the switch reluctance motor by using a hardware circuit is a direct mathematical simulation. This modelling method lays a foundation for circuit hardware modelling of the switch reluctance motor system model, and is beneficial for realizing direct mathematical simulation of a switch reluctance motor system and enables physical phenomena in the simulation system to be similar to an actual switch reluctance motor system.